Epstein-Barr virus (EBV) has major medical significance as a human tumor virus. The need for a vaccine against EBV has been recognized for 20 years, although an appropriate animal model system to study humoral and cell-mediated-immunity with good correlation to human EBV infection is lacking. We propose to develop and validate nonhuman model systems for evaluating EBV vaccine candidates. We will use this model to study and contrast the differential immunogenicity of purified gp340/220, several vaccinia-based constructs, and apathogenic EBV variants including two which will be genetically engineered. The scope of this proposal has been formulated to evaluate not only important vaccine candidates, but also to evaluate vaccine strategies. Specifically, we will: I. Perform mouse potency studies on 13 recombinant EBV vaccine candidates- Using purified EBV gp340/220 and 12 vaccinia-based recombinant vaccine candidates representing latent, early, and late EBV proteins we will perform dose potency studies in mice. II. Assess immunogenicity of selected individual EBV vaccine candidates in baboons-Based on the humoral response in mice, 6 of the 13 vaccine candidates including latent, early, and late EBV proteins will be used to determine immunogenicity in baboons, including humoral and cytotoxic T lymphocyte (CTL) responses. Based on the humoral and CTL responses in baboons of individual vaccine candidates we will test the immunogenicity of selected combinations of vaccine candidates. III. Assess immunogenicity and protection of selected vaccination regimens in common marmosets-Based on the results of immunogenicity in baboons, we will test the immunogenicity of selected vaccination regimens in common marmosets and will challenge these animals to demonstrate protection from infection with wild-type EBV. Through these tests we will validate the baboon model as a cost-efficient means of evaluating future EBV vaccine candidates prior to testing in more expensive models such as the common marmoset. IV. Assess immunogenicity and protection of genetically-engineered EBV variants in common marmosets-Using four EBV variants constructed to enhance replication and to inhibit the development of viral latency within cells, we will test immunogenicity, infectivity, viral latency, viral replication, and protection from infection with wild-type EBV in common marmosets, and compare to standard EBV and naturally-occurring EBV variants.